Good intro course, but google colab assignments need to be improved. And submitting a jupyter notebook was much more easier, why would I want to login to my google account to be a part of this course?
Great course to get started with building Convolutional Neural Networks in Keras for building Image Classifiers. This is probably the best way to get beginners into Deep Learning for Computer Vision.
par Cândido S G•
without a doubt, Laurence's teaching is much better than reading the documentation! The course is a great starting point!
par Rana T J•
The assignments need to be polished. They were very lackluster and non-rewarding.
par thomas y•
I get that this is a separate course from Ng's deep learning course, but I found the lack of theory (or even recommendations of best practices) disturbing. Additionally, I thought the videos were way too short and would have appreciated it if they had gone into detail into each Keras method used, the parameters for it, etc. For example, on the last assignment we were supposed to use a callback on accuracy to end training, but nowhere in the videos did it mention how fit_generator() handles callbacks as opposed to how they were handled with fit().
Lastly, and most importantly, this course was advertised to be a course on Tensorflow. However, this is not the case. This is a course on Keras; Tensorflow's API. If you came here looking for how to implement a DL algorithm from scratch in TF, this is not the course for you (or me apparently).
par Aniket C•
The notebooks do not provide enough information about what block implements what. A simple comment that implement custom callback here would have made things lot easier.
par Jonathan Z•
This course is generally one of high production values and the lecturer is of course excellent. Having taken Andrew Ng's original machine-learning course, it was as ever encouraging to see him make some short guest appearances.
The material and exercises are of a high standard apart from the usual typos etc. However, one particular problem was that some/all of the exercises appeared to be assessed via their print statements rather than by server-side via unit tests. This is frustrating and puts one in danger of seeing other people's code when attempting to search online for an unexpected/trial-and-error fix. Also, the timings are not quite right - some reading material comprises a few lines but apparently should take 10 minutes etc.
My main reason for undertaking the course was as a more practical prerequisite for the TensorFlow: Data and Deployment course offered by deeplearning.ai, my intention being to circle back round for the Deep Learning specialization afterwards.
Unfortunately, in spite of the positives above, the course really falls short in terms of value for money: for £45 one only learns of order 100 lines of code. This is positioned as TensorFlow having a nice, high-level API, but in practice the course is disappointingly straightforward for anyone who must have already demonstrated strong python skills and/or who are developers. There is no theory but the course certainly does not give practical "mastery" of TensorFlow.
Perhaps it was naive to expect better value, but it is hard to assess the course level purely from the syllabus. One can "audit" a course for free and I would recommend people do this in future - I certainly will be.
Overall I am quite disappointed. I look forward to the other courses in the Specialization which I am now undertaking purely in order to obtain better value for money.
par Iacopo C•
This course does an impressive job in filling the gap between theory and practice. It's part of a specialization and as such you cannot expect it to cover every single detail from the very beginning, and it's not supposed to go too deep into what Tensorflow can do.
It does an excellent job in introducing little by little the tools necessary to build neural networks, explaining the details of every line of code used and giving you everything you need to create a functioning, simple prototype.
It can't be stressed enough that this is an intermediate course and it obviously requires knowledge on the topic as well as programming in general, but it's not advanced. It's not supposed to explain the theory behind Deep Learning. It goes even as far as adding links to various free resources such as the youtube videos from Andrew Ng's Machine Learning and Deep Learning videos to help the user understand the theory behind some concepts.
par Laimonas S•
This course jumps right in and in a clear way presents a few real world like problems and solutions for them using neural networks. I am not sure how I would have felt if I had no prior ML experience taking this, it's missing a lot of basics. But perhaps that's the positive, as you get to do coding and see things in action which hopefully should energize you to dig in a bit more into things like optimizers, gradient descent, neurons, weights, training, etc. Anyway, 5 stars due to good practice examples.
par Carsten W•
I completed the 5 course Deep Learning Specialization before and was a bit unsure, if this course would still be worth it - it absolutely was. It helped me augment the theory and limited practical experience I had so far with some fun, straight to the point hints, recaps and exercises. I particularly appreciated playing with the visualization of what goes on as an image gets transformed on its way through a convnet. That was super useful to deepen the intuition.
par Jousef M•
1 week into the course so far and I really, really enjoy this course. Been following Laurence for quite a long time now and more than happy to see two of the coolest guys in AI doing such an amazing work. Although I have exams I am not missing a day watching at least 1-2 videos of them :) 100% recommended!
par disheng L•
it is a very excellent course introducing some basic knowledge about TensorFlow , teaching students how to use TensorFlow to deal with some AI programs. This course provides students with access to try their effort to use TensorFlow, and obtain more understanding about this power python library.
par Maciej Z•
This is extremely gentle intro to TensorFlow. Additionally the course provides a lot of literature including free version of Deep Learning Specialization & Machine Learning course on YT. You can easily finish it in one week, it is not time consuming at all.
par Akash D•
It is very well designed course,instructors teaches very well,i can understand all concepts and how to solve real problems using using CNN ,want such more projects based on hands on practice ,thanks
par Juan S A C•
I just can say that it was an awesome course. The instructors as well as the contents were clear, easy to understand and everything with a focus on how to take the theory and apply it with TensorFlow.
par Kristian J•
I would say that I have dabbled in Python and ML before taking this course. I even finished the IBM Data Science Professional Certification. I was very much able to grasp what was being taught here.
par Oguntoye, K•
My first course on TEnsor and i have nothing to say but this course is a must for everyone who wants to know more about Tensor. It establishes a strong background one could lay advanced knowledge on
par Mahsa G•
this course was helpful for me to learn more about CNN. it starts from basic concepts and improve that slowly and step by step. i truthfully recommend how what to start working with neural networks.
par Sebastian R G•
Es un excelente curso para todo aquel que este interesado en las ciencias de datos y la inteligencia artificial ya que Tensorflow y Keras son herrameintas que acortan mucho el camino de desarrollo.
par Sidhartha R•
It was great course to start hands-on in TensorFlow and convolution neural networks. I especially liked how the convolution feature viewing tool was used to see what the networks are learning.
par david l•
Simply Amazing Laurence! Please keep me posted on more content!!!! this is the best person to learn from he just has a way that clicks for me I hope anyone else that is taking this course!!!!
par Idan W•
As an introductory course, it filled my expectations. Short, focused videos, along with detailed notebooks. Can't wait to head on to the next course
par Muhammad I•
The course is wonderful. I happened to learn a lot. Anyone with basic skills can move on with this course pretty much easily. Highly recommended.
par Mohammed H K•
The combination of theory & practical assignments helps us in learning concepts as a beginner, and have a good learning graph.
Very useful for Beginner !
Even I have learned Deep Learning few month before, I still learn some new skills.
Thanks a lot !
par Sudhakar s•
Best in class, now I could get finally understand the basics CNN and prepared to apply DNN in realtime with confidence.
par Naveen S•
The course was very helpful in providing a essential understanding ML using tensorflow through practical steps.